@@ennucore why? 1. content - making a good piece of content requires long chain of thought and refinement. first you source good topics, then you make an outline, then you write the script, then you refine it to be more compelling and engaging, then you make it more in your tone of voice, … 2. customer support - hard tickets require deep analysis 3. resume screening - to check if the candidate is good you have to check their experience and match with job requirements, analyse how their experience plays within the context of your cultural fit
@@Oleg-Melnikov because o3 etc are RL'd for coding and math, and empirically it doesn't help them write more appealing text etc For customer support it depends - you certainly don't put o3 as the customer-facing model, but for hard tickets it should be useful For content what you need for what you described is just a good prompt/workflow, not an advanced reasoning model (it shouldn't be better than a prompt) For resume screening it again comes down to good context and workflow, e.g. extracting the text from the PDF correctly is way more important. An advanced reasoning model won't be much better at estimating cultural fit
@@ennucore I agree that for programming or maths o3 will be so much more valuable than for the resume screening, no doubt. I just wanted to show more use cases to people and to convey a general idea: "A better chain of thought will help you virtually at any task"
@@Oleg-Melnikov thanks and for real of coder with AI is not more then book that guess the word typed in line of code when we finale hit the AI AGI stage then the code will be made and processed at the same time
best time to build 🚀🚀🚀
Most people think o3 is just another incremental improvement. Nothing could be further from the truth.
Advanced reasoning models aren't really better at generating content/customer support/resume screening
@@ennucore why?
1. content - making a good piece of content requires long chain of thought and refinement. first you source good topics, then you make an outline, then you write the script, then you refine it to be more compelling and engaging, then you make it more in your tone of voice, …
2. customer support - hard tickets require deep analysis
3. resume screening - to check if the candidate is good you have to check their experience and match with job requirements, analyse how their experience plays within the context of your cultural fit
@@Oleg-Melnikov because o3 etc are RL'd for coding and math, and empirically it doesn't help them write more appealing text etc
For customer support it depends - you certainly don't put o3 as the customer-facing model, but for hard tickets it should be useful
For content what you need for what you described is just a good prompt/workflow, not an advanced reasoning model (it shouldn't be better than a prompt)
For resume screening it again comes down to good context and workflow, e.g. extracting the text from the PDF correctly is way more important. An advanced reasoning model won't be much better at estimating cultural fit
@@ennucore I agree that for programming or maths o3 will be so much more valuable than for the resume screening, no doubt. I just wanted to show more use cases to people and to convey a general idea:
"A better chain of thought will help you virtually at any task"
AI just make you for coder into youtuber hahahahahahhahhahahahahahhhahahaaaahaha
@@Lowwaels joke counts. in my opinion there is a lack of experienced software people talking on the subject, spreading it into the masses.
@@Oleg-Melnikov thanks and for real of coder with AI is not more then book that guess the word typed in line of code when we finale hit the AI AGI stage then the code will be made and processed at the same time